Overweight children have greater long-term success controlling their weight if their parents provide early support for exercising and eating right, researchers said. Ten years after participating in a weight-loss program as 6- to 12-year-olds, children whose parents got similar weight-control training were significantly less obese than those whose parents received less training, according to a study published in Wednesday's Journal of the American Medical Association. ``We think the important finding is in the modeling,'' said psychologist Leonard H. Epstein, head of the University of Pittsburgh's Childhood Obesity Clinic and leader of the research team. ``If you have children who you don't want to eat potato chips, you can't sit in front of them eating Ho-Hos.'' Epstein's group also concluded that pre-adolescent weight loss had no effect on height development, contrary to the findings of several shorter-term studies. Two other childhood obesity experts called Epstein's results surprisin because previous research has shown that fat children will grow into fat adults and that weight lost in treatment programs is quickly regained. Those experts said further study was needed on whether treatment of obesity in childhood can produce effects that persist into young adulthood. In Epstein's project, an initial group of 75 obese children and their parents were randomly assigned to three groups. Study criteria required the children be more than 20 percent over ideal weight for their age, height and sex; both parents living at home, one of whom had to be obese; and one parent willing to attend treatment meetings with their child. All subjects were to attend eight weekly treatment meetings followed by six monthly meetings, then by 21-month, five-year and 10-year meetings. All were instructed to follow a regimen that included a 1,200-to-1,500-calorie-per-day diet and an aerobic exercise program. In the first group, both parents and children received reinforcement for losing weight and were trained to praise other family members for changing eating and exercise habits. In the second group, only the children received reinforcement for losing weight but were similarly trained to praise family members for weight-control behaviors. Parents and children in the third group received reinforcement only for attending the meetings. They received information about social reinforcement without specific training in how to use the method. All subjects lost weight initially. At the 10-year mark, the children in the first group were, on average, 7.5 percent less overweight than at the beginning of the study while those in the second and third groups were 4.5 percent and 14.3 percent fatter, respectively. Parents in all three groups were more obese after 10 years than at the start of the study. In a journal editorial, Dr. Albert J. Stunkard of the University of Pennsylvania medical school and Dr. Robert I. Berkowitz of the Philadelphia Child Guidance Clinic cautioned that the ``treatment did not prevent obesity, it limited it.'' They called Epstein's results surprising, perhaps because of his relatively small study sample: intact, middle-class families in which the parents were likely to be very supportive of their children's efforts. They said replication of Epstein's study ``is clearly needed'' before the findings can be embraced.